INDEX

Explanations

statements related to statistics and records

oai_token-act-pair · gpt-3.5-turbo

phrases related to significant metrics or statistics regarding confidence and approval ratings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

akia

-0.67

 surn

-0.65

nington

-0.65

 faithfully

-0.64

 WATCHED

-0.64

 congratulated

-0.64

sbm

-0.63

pard

-0.63

quartered

-0.62

 resemb

-0.62

POSITIVE LOGITS

 lows

1.07

 highs

1.02

 levels

0.95

 peak

0.93

 peaks

0.93

risis

0.89

 heights

0.89

 lowest

0.86

Hig

0.83

 saturation

0.81

Activations Density 0.271%

statements related to statistics and records

phrases related to significant metrics or statistics regarding confidence and approval ratings

No Comments

No Known Activations